Windows 11: New pushed ip address not applied on seamless reconnect

This forum is for admins who are looking to build or expand their OpenVPN setup.
Forum rules
Please use the [oconf] BB tag for openvpn Configurations. See viewtopic.php?f=30&t=21589 for an example.
Post Reply
OpenVpn Newbie
Posts: 4
Joined: Tue Jan 03, 2023 10:22 am

Windows 11: New pushed ip address not applied on seamless reconnect

Post by basinilya » Tue Jun 06, 2023 8:49 am

Sorry, I haven't found the forum branch to discuss the Client.

I recently migrated to a fresh Windows 11 and noticed that after resuming from long sleep there's no traffic for hours until I manually reconnect. According to the client log it performs a seamless reconnect after sleep and then it thinks that everything is fine (nothing is logged).

I found a line that was logged right after resuming:

Code: Select all

PUSH: Received control message: 'PUSH_REPLY,...,ifconfig,...'
and the address was different from the address assigned to my virtual adapter.
After I assigned the new address manually with "netsh interface ipv4 add address" I was able to reach the hosts on the other side of the tunnel. Nothing was logged by the client during the process so the client didn't even notice the change.

So why did it start happening only recently? I see these differences on the fresh Windows 11 :
  • the client does not install a tun/tap adapter. It installs some "OpenVPN Data Channel Offload" which I never heard of before
  • the client does not assign a static ip address! It keeps the adapter in the "dhcp" mode even though there's no DHCP on the other side
  • I believe (though I have no proof) the client calls the "AddIPAddress" API to add the pushed address to the adapter, but it does not call "DeleteIPAddress" on seamless reconnect and that must be the bug.
Unfortunately, "netsh interface ipv4 add address" puts the adapter into manual mode so it's not the same thing.

I use OpenVPN 2.6.3-I001 amd64 with OpenVPN Gui 11.41


Post Reply